In [1]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

Exploring data

In [2]:
zomatodata = pd.read_csv(r"D:\Projects\EDA Zomato\zomato_restaurants_in_India.csv")
In [3]:
pd.set_option('display.max_columns', None)
In [4]:
zomatodata
Out[4]:
res_id name establishment url address city city_id locality latitude longitude zipcode country_id locality_verbose cuisines timings average_cost_for_two price_range currency highlights aggregate_rating rating_text votes photo_count opentable_support delivery takeaway
0 3400299 Bikanervala ['Quick Bites'] https://www.zomato.com/agra/bikanervala-khanda... Kalyani Point, Near Tulsi Cinema, Bypass Road,... Agra 34 Khandari 27.211450 78.002381 NaN 1 Khandari, Agra North Indian, South Indian, Mithai, Street Foo... 8:30am – 10:30pm (Mon-Sun) 700 2 Rs. ['Lunch', 'Takeaway Available', 'Credit Card',... 4.4 Very Good 814 154 0.0 -1 -1
1 3400005 Mama Chicken Mama Franky House ['Quick Bites'] https://www.zomato.com/agra/mama-chicken-mama-... Main Market, Sadar Bazaar, Agra Cantt, Agra Agra 34 Agra Cantt 27.160569 78.011583 282001.0 1 Agra Cantt, Agra North Indian, Mughlai, Rolls, Chinese, Fast Fo... 12:30PM to 12Midnight (Mon, Wed, Thu, Fri, Sat... 600 2 Rs. ['Delivery', 'No Alcohol Available', 'Dinner',... 4.4 Very Good 1203 161 0.0 -1 -1
2 3401013 Bhagat Halwai ['Quick Bites'] https://www.zomato.com/agra/bhagat-halwai-2-sh... 62/1, Near Easy Day, West Shivaji Nagar, Goalp... Agra 34 Shahganj 27.182938 77.979684 282010.0 1 Shahganj, Agra Fast Food, Mithai 9:30 AM to 11 PM 300 1 Rs. ['No Alcohol Available', 'Dinner', 'Takeaway A... 4.2 Very Good 801 107 0.0 1 -1
3 3400290 Bhagat Halwai ['Quick Bites'] https://www.zomato.com/agra/bhagat-halwai-civi... Near Anjana Cinema, Nehru Nagar, Civil Lines, ... Agra 34 Civil Lines 27.205668 78.004799 282002.0 1 Civil Lines, Agra Desserts, Bakery, Fast Food, South Indian 8am – 11pm (Mon-Sun) 300 1 Rs. ['Takeaway Available', 'Credit Card', 'Lunch',... 4.3 Very Good 693 157 0.0 1 -1
4 3401744 The Salt Cafe Kitchen & Bar ['Casual Dining'] https://www.zomato.com/agra/the-salt-cafe-kitc... 1C,3rd Floor, Fatehabad Road, Tajganj, Agra Agra 34 Tajganj 27.157709 78.052421 NaN 1 Tajganj, Agra North Indian, Continental, Italian 11:30 AM to 11:30 PM 1000 3 Rs. ['Lunch', 'Serves Alcohol', 'Cash', 'Credit Ca... 4.9 Excellent 470 291 0.0 1 -1
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
211939 3202251 Kali Mirch Cafe And Restaurant ['Casual Dining'] https://www.zomato.com/vadodara/kali-mirch-caf... Manu Smriti Complex, Near Navrachna School, GI... Vadodara 32 Fatehgunj 22.336931 73.192356 390024.0 1 Fatehgunj, Vadodara North Indian 11am – 3pm, 7pm – 11:30pm (Mon-Sun) 600 2 Rs. ['Dinner', 'Cash', 'Lunch', 'Delivery', 'Indoo... 4.1 Very Good 243 40 0.0 -1 -1
211940 3200996 Raju Omlet ['Quick Bites'] https://www.zomato.com/vadodara/raju-omlet-kar... Mahalaxmi Apartment, Opposite B O B, Karoli Ba... Vadodara 32 Karelibaug 22.322455 73.197203 NaN 1 Karelibaug, Vadodara Fast Food 4pm – 11pm (Mon, Tue, Wed, Fri, Sat, Sun), 4pm... 300 1 Rs. ['Dinner', 'Cash', 'Takeaway Available', 'Debi... 4.1 Very Good 187 40 0.0 1 -1
211941 18984164 The Grand Thakar ['Casual Dining'] https://www.zomato.com/vadodara/the-grand-thak... 3rd Floor, Shreem Shalini Mall, Opposite Conqu... Vadodara 32 Alkapuri 22.310563 73.171163 NaN 1 Alkapuri, Vadodara Gujarati, North Indian, Chinese 11:30 AM to 3:30 PM, 7:30 PM to 11 PM 700 2 Rs. ['Dinner', 'Cash', 'Debit Card', 'Lunch', 'Tak... 4.0 Very Good 111 38 0.0 -1 -1
211942 3201138 Subway ['Quick Bites'] https://www.zomato.com/vadodara/subway-1-akota... G-2, Vedant Platina, Near Cosmos, Akota, Vadodara Vadodara 32 Akota 22.270027 73.143068 NaN 1 Akota, Vadodara Fast Food, Sandwich, Salad 8 AM to 1 AM 500 2 Rs. ['Dinner', 'Delivery', 'Credit Card', 'Lunch',... 3.7 Good 128 34 0.0 1 -1
211943 18879846 Freshco's - The Health Cafe ['Café'] https://www.zomato.com/vadodara/freshcos-the-h... Shop 7, Ground Floor, Opposite Natubhai Circle... Vadodara 32 Vadiwadi 22.309935 73.158768 390007.0 1 Vadiwadi, Vadodara Cafe, Healthy Food, Coffee 7am – 11pm (Mon-Sun) 600 2 Rs. ['Dinner', 'Cash', 'Takeaway Available', 'Debi... 4.0 Very Good 93 53 0.0 1 -1

211944 rows × 26 columns

In [5]:
zomatodata.shape
Out[5]:
(211944, 26)

The dataset got 26 columns and 211944 rows.

In [6]:
zomatodata.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 211944 entries, 0 to 211943
Data columns (total 26 columns):
 #   Column                Non-Null Count   Dtype  
---  ------                --------------   -----  
 0   res_id                211944 non-null  int64  
 1   name                  211944 non-null  object 
 2   establishment         211944 non-null  object 
 3   url                   211944 non-null  object 
 4   address               211810 non-null  object 
 5   city                  211944 non-null  object 
 6   city_id               211944 non-null  int64  
 7   locality              211944 non-null  object 
 8   latitude              211944 non-null  float64
 9   longitude             211944 non-null  float64
 10  zipcode               48757 non-null   object 
 11  country_id            211944 non-null  int64  
 12  locality_verbose      211944 non-null  object 
 13  cuisines              210553 non-null  object 
 14  timings               208070 non-null  object 
 15  average_cost_for_two  211944 non-null  int64  
 16  price_range           211944 non-null  int64  
 17  currency              211944 non-null  object 
 18  highlights            211944 non-null  object 
 19  aggregate_rating      211944 non-null  float64
 20  rating_text           211944 non-null  object 
 21  votes                 211944 non-null  int64  
 22  photo_count           211944 non-null  int64  
 23  opentable_support     211896 non-null  float64
 24  delivery              211944 non-null  int64  
 25  takeaway              211944 non-null  int64  
dtypes: float64(4), int64(9), object(13)
memory usage: 42.0+ MB
In [7]:
zomatodata.describe()
Out[7]:
res_id city_id latitude longitude country_id average_cost_for_two price_range aggregate_rating votes photo_count opentable_support delivery takeaway
count 2.119440e+05 211944.000000 211944.000000 211944.000000 211944.0 211944.000000 211944.000000 211944.000000 211944.000000 211944.000000 211896.0 211944.000000 211944.0
mean 1.349411e+07 4746.785434 21.499758 77.615276 1.0 595.812229 1.882535 3.395937 378.001864 256.971224 0.0 -0.255907 -1.0
std 7.883722e+06 5568.766386 22.781331 7.500104 0.0 606.239363 0.892989 1.283642 925.333370 867.668940 0.0 0.964172 0.0
min 5.000000e+01 1.000000 0.000000 0.000000 1.0 0.000000 1.000000 0.000000 -18.000000 0.000000 0.0 -1.000000 -1.0
25% 3.301027e+06 11.000000 15.496071 74.877961 1.0 250.000000 1.000000 3.300000 16.000000 3.000000 0.0 -1.000000 -1.0
50% 1.869573e+07 34.000000 22.514494 77.425971 1.0 400.000000 2.000000 3.800000 100.000000 18.000000 0.0 -1.000000 -1.0
75% 1.881297e+07 11306.000000 26.841667 80.219323 1.0 700.000000 2.000000 4.100000 362.000000 128.000000 0.0 1.000000 -1.0
max 1.915979e+07 11354.000000 10000.000000 91.832769 1.0 30000.000000 4.000000 4.900000 42539.000000 17702.000000 0.0 1.000000 -1.0

Data Cleaning and Preparation:¶

  • Identify and handle missing values.
  • Detect and correct any inconsistencies in the dataset (e.g., data types, mislabeled categories).
  • Feature engineering (if necessary), like extracting useful information from existing data.

Checkin for duplicates and removing them.¶

As 'res_id' is a unique identifier for the restaurants, we will use it to identify and remove duplicates.

In [8]:
zomatodata["res_id"].duplicated().sum()
Out[8]:
156376

total rows = 211944\ duplicates rows = 156376\

  • 73% of our data has duplicate rows. Removing this data will help in performing better analysis and avoid any biasness.
In [9]:
zomatodata.drop_duplicates(["res_id"], keep='first', inplace = True )
In [10]:
zomatodata.shape
Out[10]:
(55568, 26)

Total rows after removing duplicates.\ Total rows = 55568

Copying Dataframe¶

  • Making a copy of original dataframe before applying EDA
In [11]:
zomato = zomatodata.copy()
zomato.head()
Out[11]:
res_id name establishment url address city city_id locality latitude longitude zipcode country_id locality_verbose cuisines timings average_cost_for_two price_range currency highlights aggregate_rating rating_text votes photo_count opentable_support delivery takeaway
0 3400299 Bikanervala ['Quick Bites'] https://www.zomato.com/agra/bikanervala-khanda... Kalyani Point, Near Tulsi Cinema, Bypass Road,... Agra 34 Khandari 27.211450 78.002381 NaN 1 Khandari, Agra North Indian, South Indian, Mithai, Street Foo... 8:30am – 10:30pm (Mon-Sun) 700 2 Rs. ['Lunch', 'Takeaway Available', 'Credit Card',... 4.4 Very Good 814 154 0.0 -1 -1
1 3400005 Mama Chicken Mama Franky House ['Quick Bites'] https://www.zomato.com/agra/mama-chicken-mama-... Main Market, Sadar Bazaar, Agra Cantt, Agra Agra 34 Agra Cantt 27.160569 78.011583 282001.0 1 Agra Cantt, Agra North Indian, Mughlai, Rolls, Chinese, Fast Fo... 12:30PM to 12Midnight (Mon, Wed, Thu, Fri, Sat... 600 2 Rs. ['Delivery', 'No Alcohol Available', 'Dinner',... 4.4 Very Good 1203 161 0.0 -1 -1
2 3401013 Bhagat Halwai ['Quick Bites'] https://www.zomato.com/agra/bhagat-halwai-2-sh... 62/1, Near Easy Day, West Shivaji Nagar, Goalp... Agra 34 Shahganj 27.182938 77.979684 282010.0 1 Shahganj, Agra Fast Food, Mithai 9:30 AM to 11 PM 300 1 Rs. ['No Alcohol Available', 'Dinner', 'Takeaway A... 4.2 Very Good 801 107 0.0 1 -1
3 3400290 Bhagat Halwai ['Quick Bites'] https://www.zomato.com/agra/bhagat-halwai-civi... Near Anjana Cinema, Nehru Nagar, Civil Lines, ... Agra 34 Civil Lines 27.205668 78.004799 282002.0 1 Civil Lines, Agra Desserts, Bakery, Fast Food, South Indian 8am – 11pm (Mon-Sun) 300 1 Rs. ['Takeaway Available', 'Credit Card', 'Lunch',... 4.3 Very Good 693 157 0.0 1 -1
4 3401744 The Salt Cafe Kitchen & Bar ['Casual Dining'] https://www.zomato.com/agra/the-salt-cafe-kitc... 1C,3rd Floor, Fatehabad Road, Tajganj, Agra Agra 34 Tajganj 27.157709 78.052421 NaN 1 Tajganj, Agra North Indian, Continental, Italian 11:30 AM to 11:30 PM 1000 3 Rs. ['Lunch', 'Serves Alcohol', 'Cash', 'Credit Ca... 4.9 Excellent 470 291 0.0 1 -1
In [12]:
zomato.shape
Out[12]:
(55568, 26)
In [13]:
zomato.info()

<class 'pandas.core.frame.DataFrame'>
Index: 55568 entries, 0 to 211942
Data columns (total 26 columns):
 #   Column                Non-Null Count  Dtype  
---  ------                --------------  -----  
 0   res_id                55568 non-null  int64  
 1   name                  55568 non-null  object 
 2   establishment         55568 non-null  object 
 3   url                   55568 non-null  object 
 4   address               55550 non-null  object 
 5   city                  55568 non-null  object 
 6   city_id               55568 non-null  int64  
 7   locality              55568 non-null  object 
 8   latitude              55568 non-null  float64
 9   longitude             55568 non-null  float64
 10  zipcode               10945 non-null  object 
 11  country_id            55568 non-null  int64  
 12  locality_verbose      55568 non-null  object 
 13  cuisines              55098 non-null  object 
 14  timings               54565 non-null  object 
 15  average_cost_for_two  55568 non-null  int64  
 16  price_range           55568 non-null  int64  
 17  currency              55568 non-null  object 
 18  highlights            55568 non-null  object 
 19  aggregate_rating      55568 non-null  float64
 20  rating_text           55568 non-null  object 
 21  votes                 55568 non-null  int64  
 22  photo_count           55568 non-null  int64  
 23  opentable_support     55556 non-null  float64
 24  delivery              55568 non-null  int64  
 25  takeaway              55568 non-null  int64  
dtypes: float64(4), int64(9), object(13)
memory usage: 11.4+ MB

Striping column names.¶

In [14]:
zomato.columns.str.strip()
Out[14]:
Index(['res_id', 'name', 'establishment', 'url', 'address', 'city', 'city_id',
       'locality', 'latitude', 'longitude', 'zipcode', 'country_id',
       'locality_verbose', 'cuisines', 'timings', 'average_cost_for_two',
       'price_range', 'currency', 'highlights', 'aggregate_rating',
       'rating_text', 'votes', 'photo_count', 'opentable_support', 'delivery',
       'takeaway'],
      dtype='object')

Column names.¶

In [15]:
colname = zomato.columns
for i in range(len(colname)):
    print(str(i)+"."+colname[i])

0.res_id
1.name
2.establishment
3.url
4.address
5.city
6.city_id
7.locality
8.latitude
9.longitude
10.zipcode
11.country_id
12.locality_verbose
13.cuisines
14.timings
15.average_cost_for_two
16.price_range
17.currency
18.highlights
19.aggregate_rating
20.rating_text
21.votes
22.photo_count
23.opentable_support
24.delivery
25.takeaway

We have 26 columns in the dataframe. We can check each each column and decide weather we need it for our analysis.

Checkin missing values.¶

In [16]:
zomato.isna().sum()
Out[16]:
res_id                      0
name                        0
establishment               0
url                         0
address                    18
city                        0
city_id                     0
locality                    0
latitude                    0
longitude                   0
zipcode                 44623
country_id                  0
locality_verbose            0
cuisines                  470
timings                  1003
average_cost_for_two        0
price_range                 0
currency                    0
highlights                  0
aggregate_rating            0
rating_text                 0
votes                       0
photo_count                 0
opentable_support          12
delivery                    0
takeaway                    0
dtype: int64

There 5 columns which have missing values.

  • address = 18
  • zipcode = 47869
  • cuisines = 470
  • timings = 1070
  • opentable_support = 19

Check columns data types.¶

In [17]:
zomato.dtypes
Out[17]:
res_id                    int64
name                     object
establishment            object
url                      object
address                  object
city                     object
city_id                   int64
locality                 object
latitude                float64
longitude               float64
zipcode                  object
country_id                int64
locality_verbose         object
cuisines                 object
timings                  object
average_cost_for_two      int64
price_range               int64
currency                 object
highlights               object
aggregate_rating        float64
rating_text              object
votes                     int64
photo_count               int64
opentable_support       float64
delivery                  int64
takeaway                  int64
dtype: object

establishment¶

In [18]:
zomato['establishment'].unique()
Out[18]:
array(["['Quick Bites']", "['Casual Dining']", "['Bakery']", "['Café']",
       "['Dhaba']", "['Bhojanalya']", "['Bar']", "['Sweet Shop']",
       "['Fine Dining']", "['Food Truck']", "['Dessert Parlour']",
       "['Lounge']", "['Pub']", "['Beverage Shop']", "['Kiosk']",
       "['Paan Shop']", "['Confectionery']", '[]', "['Shack']",
       "['Club']", "['Food Court']", "['Mess']", "['Butcher Shop']",
       "['Microbrewery']", "['Cocktail Bar']", "['Pop up']",
       "['Irani Cafe']"], dtype=object)
  • Removin the unnessary barckets([,]) and quotes(').
  • Replacing the empty string([]) with 'NA'= Not Avalibe.
  • changin 'Café' to 'cafe.
In [19]:
zomato["establishment"] = zomato["establishment"].apply(lambda x:x[2:-2])

zomato['establishment'] = zomato['establishment'].apply(lambda x: np.where(x=="", "Not Avalible", x))
In [20]:
zomato['establishment'].unique()
Out[20]:
array(['Quick Bites', 'Casual Dining', 'Bakery', 'Café', 'Dhaba',
       'Bhojanalya', 'Bar', 'Sweet Shop', 'Fine Dining', 'Food Truck',
       'Dessert Parlour', 'Lounge', 'Pub', 'Beverage Shop', 'Kiosk',
       'Paan Shop', 'Confectionery', 'Not Avalible', 'Shack', 'Club',
       'Food Court', 'Mess', 'Butcher Shop', 'Microbrewery',
       'Cocktail Bar', 'Pop up', 'Irani Cafe'], dtype=object)
In [21]:
zomato['establishment'].value_counts()
Out[21]:
establishment
Quick Bites        14032
Casual Dining      12270
Café                4123
Bakery              3741
Dessert Parlour     3675
Sweet Shop          2615
Beverage Shop       2440
Not Avalible        1830
Fine Dining         1535
Food Court          1494
Bar                 1399
Dhaba               1282
Kiosk               1126
Food Truck           868
Lounge               820
Bhojanalya           632
Mess                 361
Pub                  357
Paan Shop            320
Confectionery        218
Butcher Shop         154
Club                 112
Microbrewery         110
Shack                 20
Cocktail Bar          16
Irani Cafe            14
Pop up                 4
Name: count, dtype: int64

address¶

  • Missin 18 vlues.
  • Filling missing value with Placeholder 'No address provided'.
In [22]:
zomato['address'].fillna('No address provided', inplace=True)
In [23]:
zomato['address'].isna().sum()
Out[23]:
0

city¶

In [24]:
zomato['city'].unique()
Out[24]:
array(['Agra', 'Ahmedabad', 'Gandhinagar', 'Ajmer', 'Alappuzha',
       'Allahabad', 'Amravati', 'Amritsar', 'Aurangabad', 'Bangalore',
       'Bhopal', 'Bhubaneshwar', 'Chandigarh', 'Mohali', 'Panchkula',
       'Zirakpur', 'Nayagaon', 'Chennai', 'Coimbatore', 'Cuttack',
       'Darjeeling', 'Dehradun', 'New Delhi', 'Gurgaon', 'Noida',
       'Faridabad', 'Ghaziabad', 'Greater Noida', 'Dharamshala',
       'Gangtok', 'Goa', 'Gorakhpur', 'Guntur', 'Guwahati', 'Gwalior',
       'Haridwar', 'Hyderabad', 'Secunderabad', 'Indore', 'Jabalpur',
       'Jaipur', 'Jalandhar', 'Jammu', 'Jamnagar', 'Jamshedpur', 'Jhansi',
       'Jodhpur', 'Junagadh', 'Kanpur', 'Kharagpur', 'Kochi', 'Kolhapur',
       'Kolkata', 'Howrah', 'Kota', 'Lucknow', 'Ludhiana', 'Madurai',
       'Manali', 'Mangalore', 'Manipal', 'Udupi', 'Meerut', 'Mumbai',
       'Thane', 'Navi Mumbai', 'Mussoorie', 'Mysore', 'Nagpur',
       'Nainital', 'Nasik', 'Nashik', 'Neemrana', 'Ooty', 'Palakkad',
       'Patiala', 'Patna', 'Puducherry', 'Pune', 'Pushkar', 'Raipur',
       'Rajkot', 'Ranchi', 'Rishikesh', 'Salem', 'Shimla', 'Siliguri',
       'Srinagar', 'Surat', 'Thrissur', 'Tirupati', 'Trichy',
       'Trivandrum', 'Udaipur', 'Varanasi', 'Vellore', 'Vijayawada',
       'Vizag', 'Vadodara'], dtype=object)
In [25]:
len(zomato['city'].unique())
Out[25]:
99
In [26]:
zomato['city'].value_counts()
Out[26]:
city
Bangalore        2247
Mumbai           2022
Pune             1843
Chennai          1827
New Delhi        1704
                 ... 
Udupi              60
Howrah             50
Neemrana           26
Greater Noida      21
Nayagaon           15
Name: count, Length: 99, dtype: int64

city_id¶

In [27]:
len(zomato['city_id'].unique())
Out[27]:
83

locality¶

In [28]:
zomato['locality'].nunique()
Out[28]:
3731

zipcode¶

  • It is missing 86% of the data and is not relevant for the current analysis, so we can remove it the the data.
In [29]:
zomato.drop(columns=['zipcode'], inplace=True)

country_id¶

In [30]:
zomato['country_id'].unique()
Out[30]:
array([1], dtype=int64)

locality_verbose¶

In [31]:
zomato["locality_verbose"].nunique()
Out[31]:
3910

cuisines¶

Contain 470 missing vaules.

In [32]:
zomato["cuisines"].nunique()
Out[32]:
9382
In [33]:
zomato["cuisines"].unique()
Out[33]:
array(['North Indian, South Indian, Mithai, Street Food, Desserts',
       'North Indian, Mughlai, Rolls, Chinese, Fast Food, Street Food',
       'Fast Food, Mithai', ...,
       'Street Food, Biryani, Chinese, Fast Food, North Indian, Mughlai',
       'North Indian, Chinese, Mexican, Italian, Thai, Continental',
       'North Indian, Lucknowi, Chinese'], dtype=object)
In [34]:
zomato['cuisines'] = zomato['cuisines'].fillna('Not Specified')
In [35]:
zomato['cuisines'].isna().sum()
Out[35]:
0

Extracting and analyzing the unique cuisines listed in a DataFrame

In [36]:
cuisines = []
zomato['cuisines'].apply(lambda x : cuisines.extend(x.split(', ')))
cuisines = pd.Series(cuisines)
cuisines.nunique()
Out[36]:
134

Timings¶

It is missing 1003 values. It is also not structured and is not very usefull for the analysis so we can remove it.

In [37]:
zomato['timings'].isna().sum()
Out[37]:
1003
In [38]:
zomato['timings'].unique()
Out[38]:
array(['8:30am – 10:30pm (Mon-Sun)',
       '12:30PM to 12Midnight (Mon, Wed, Thu, Fri, Sat, Sun), 1PM to 12Midnight (Tue)',
       '9:30 AM to 11 PM', ..., '8am – 2:30pm, 5pm – 10pm (Mon-Sun)',
       '8am – 3pm, 6:30pm – 11pm (Mon-Sun)',
       '4pm – 11pm (Mon, Tue, Wed, Fri, Sat, Sun), 4pm – 11:30pm (Thu)'],
      dtype=object)
In [39]:
zomato['timings'].nunique()
Out[39]:
7740

average_cost_for_two¶

In [40]:
zomato["average_cost_for_two"].nunique()
Out[40]:
145

price_range¶

In [41]:
zomato['price_range'].unique()
Out[41]:
array([2, 1, 3, 4], dtype=int64)

highlights¶

In [42]:
zomato['highlights'].unique()
Out[42]:
array(["['Lunch', 'Takeaway Available', 'Credit Card', 'Dinner', 'Cash', 'Air Conditioned', 'Indoor Seating', 'Pure Veg']",
       "['Delivery', 'No Alcohol Available', 'Dinner', 'Takeaway Available', 'Lunch', 'Cash', 'Indoor Seating']",
       "['No Alcohol Available', 'Dinner', 'Takeaway Available', 'Breakfast', 'Lunch', 'Cash', 'Delivery', 'Outdoor Seating', 'Air Conditioned', 'Self Service', 'Indoor Seating', 'Digital Payments Accepted', 'Pure Veg', 'Desserts and Bakes']",
       ...,
       "['Dinner', 'Delivery', 'Cash', 'Takeaway Available', 'Free Parking', 'Digital Payments Accepted', 'Pure Veg', 'Indoor Seating']",
       "['Dinner', 'Cash', 'Takeaway Available', 'Lunch', 'Delivery', 'Free Parking', 'Indoor Seating', 'Air Conditioned', 'Outdoor Seating', 'Digital Payments Accepted', 'Catering Available', 'Pure Veg']",
       "['Dinner', 'Cash', 'Takeaway Available', 'Debit Card', 'Delivery', 'Credit Card', 'Free Parking', 'Outdoor Seating']"],
      dtype=object)
In [43]:
zomato['highlights'].nunique()
Out[43]:
31228
In [44]:
highlight = []
zomato["highlights"].apply(lambda x : highlight.extend(x[2:-2].split("', '")))
highlight = pd.Series(highlight)
highlight.nunique()
Out[44]:
104
In [45]:
zomato['highlights']
Out[45]:
0         ['Lunch', 'Takeaway Available', 'Credit Card',...
1         ['Delivery', 'No Alcohol Available', 'Dinner',...
2         ['No Alcohol Available', 'Dinner', 'Takeaway A...
3         ['Takeaway Available', 'Credit Card', 'Lunch',...
4         ['Lunch', 'Serves Alcohol', 'Cash', 'Credit Ca...
                                ...                        
211882    ['Cash', 'Takeaway Available', 'Delivery', 'In...
211925    ['Dinner', 'Cash', 'Debit Card', 'Lunch', 'Tak...
211926    ['Dinner', 'Cash', 'Credit Card', 'Lunch', 'Ta...
211940    ['Dinner', 'Cash', 'Takeaway Available', 'Debi...
211942    ['Dinner', 'Delivery', 'Credit Card', 'Lunch',...
Name: highlights, Length: 55568, dtype: object

rating_text¶

In [46]:
zomato['rating_text'].unique()
Out[46]:
array(['Very Good', 'Excellent', 'Good', 'Average', 'Not rated', 'Poor',
       'Bardzo dobrze', 'Muito Bom', 'İyi', 'Çok iyi', 'Baik',
       'Sangat Baik', 'Skvělé', 'Muy Bueno', 'Průměr', 'Ortalama',
       'Skvělá volba', 'Muito bom', 'Velmi dobré', 'Excelente', 'Bom',
       'Scarso', 'Promedio', 'Vynikajúce', 'Bueno', 'Dobré', 'Terbaik',
       'Harika', 'Veľmi dobré', 'Eccellente', 'Buono', 'Média', 'Dobrze'],
      dtype=object)

Mapping non-english word and converting it to english by using dictonary.

In [47]:
rating_translation = {
    'Dobré': 'Good', 'Baik': 'Good', 'Sangat Baik': 'Very Good',
    'Excelente': 'Excellent', 'Bardzo dobrze': 'Very Good',
    'Wybitnie': 'Excellent', 'Ottimo': 'Excellent', 'Muito Bom': 'Very Good',
    'Velmi dobré': 'Very Good', 'Skvělá volba': 'Excellent',
    'Muy Bueno': 'Very Good', 'Bom': 'Good', 'İyi': 'Good', 'Çok iyi': 'Very Good',
    'Harika': 'Excellent', 'Terbaik': 'Excellent', 'Skvělé': 'Excellent', 'Průměr': 'Average',
    'Ortalama': 'Average', 'Bueno': 'Good', 'Eccellente': 'Excellent',
    'Muito bom': 'Very Good', 'Dobrze': 'Good', 'Buono': 'Good', 'Média': 'Average',
    'Scarso': 'Poor', 'Promedio': 'Average', 'Veľmi dobré': 'Very Good',
    'Vynikajúce': 'Excellent', 'Średnio': 'Average', 'Priemer': 'Average',
    'Biasa': 'Average', 'Media': 'Average'
}

zomato['rating_text'] = zomato['rating_text'].replace(rating_translation)

zomato['rating_text'].unique()
Out[47]:
array(['Very Good', 'Excellent', 'Good', 'Average', 'Not rated', 'Poor'],
      dtype=object)

opentable_support¶

missing 12 vulues, replacing missing with 0

In [48]:
zomato['opentable_support'].unique()
Out[48]:
array([ 0., nan])
In [49]:
zomato['opentable_support'].fillna(0, inplace=True)
In [50]:
zomato['opentable_support'].unique()
Out[50]:
array([0.])

delivery¶

In [51]:
zomato["delivery"].unique()
Out[51]:
array([-1,  1,  0], dtype=int64)

takeaway¶

In [52]:
zomato["takeaway"].unique()
Out[52]:
array([-1], dtype=int64)
In [53]:
zomato.head()
Out[53]:
res_id name establishment url address city city_id locality latitude longitude country_id locality_verbose cuisines timings average_cost_for_two price_range currency highlights aggregate_rating rating_text votes photo_count opentable_support delivery takeaway
0 3400299 Bikanervala Quick Bites https://www.zomato.com/agra/bikanervala-khanda... Kalyani Point, Near Tulsi Cinema, Bypass Road,... Agra 34 Khandari 27.211450 78.002381 1 Khandari, Agra North Indian, South Indian, Mithai, Street Foo... 8:30am – 10:30pm (Mon-Sun) 700 2 Rs. ['Lunch', 'Takeaway Available', 'Credit Card',... 4.4 Very Good 814 154 0.0 -1 -1
1 3400005 Mama Chicken Mama Franky House Quick Bites https://www.zomato.com/agra/mama-chicken-mama-... Main Market, Sadar Bazaar, Agra Cantt, Agra Agra 34 Agra Cantt 27.160569 78.011583 1 Agra Cantt, Agra North Indian, Mughlai, Rolls, Chinese, Fast Fo... 12:30PM to 12Midnight (Mon, Wed, Thu, Fri, Sat... 600 2 Rs. ['Delivery', 'No Alcohol Available', 'Dinner',... 4.4 Very Good 1203 161 0.0 -1 -1
2 3401013 Bhagat Halwai Quick Bites https://www.zomato.com/agra/bhagat-halwai-2-sh... 62/1, Near Easy Day, West Shivaji Nagar, Goalp... Agra 34 Shahganj 27.182938 77.979684 1 Shahganj, Agra Fast Food, Mithai 9:30 AM to 11 PM 300 1 Rs. ['No Alcohol Available', 'Dinner', 'Takeaway A... 4.2 Very Good 801 107 0.0 1 -1
3 3400290 Bhagat Halwai Quick Bites https://www.zomato.com/agra/bhagat-halwai-civi... Near Anjana Cinema, Nehru Nagar, Civil Lines, ... Agra 34 Civil Lines 27.205668 78.004799 1 Civil Lines, Agra Desserts, Bakery, Fast Food, South Indian 8am – 11pm (Mon-Sun) 300 1 Rs. ['Takeaway Available', 'Credit Card', 'Lunch',... 4.3 Very Good 693 157 0.0 1 -1
4 3401744 The Salt Cafe Kitchen & Bar Casual Dining https://www.zomato.com/agra/the-salt-cafe-kitc... 1C,3rd Floor, Fatehabad Road, Tajganj, Agra Agra 34 Tajganj 27.157709 78.052421 1 Tajganj, Agra North Indian, Continental, Italian 11:30 AM to 11:30 PM 1000 3 Rs. ['Lunch', 'Serves Alcohol', 'Cash', 'Credit Ca... 4.9 Excellent 470 291 0.0 1 -1

Exploratory Data Analysis:¶

  • Descriptive Statistics: Summarize the central tendency, dispersion, and shape of the dataset's distribution.
  • Distribution Analysis: Analyze the distribution of key variables (e.g., ratings, price range, cuisines).
  • Correlation Analysis: Examine the relationships between different variables.

Descriptive Statistics¶

In [54]:
zomato.shape
Out[54]:
(55568, 25)
In [55]:
summary_stats = zomato.describe()
summary_stats
Out[55]:
res_id city_id latitude longitude country_id average_cost_for_two price_range aggregate_rating votes photo_count opentable_support delivery takeaway
count 5.556800e+04 55568.000000 55568.000000 55568.000000 55568.0 55568.000000 55568.000000 55568.000000 55568.000000 55568.000000 55568.0 55568.000000 55568.0
mean 1.313694e+07 3409.499298 21.450847 76.497131 1.0 528.205874 1.714728 2.958593 223.330352 160.974770 0.0 -0.349824 -1.0
std 8.105959e+06 5174.942737 42.901135 10.982976 0.0 595.028447 0.878227 1.464576 618.224019 586.975382 0.0 0.933794 0.0
min 5.000000e+01 1.000000 0.000000 0.000000 1.0 0.000000 1.000000 0.000000 -18.000000 0.000000 0.0 -1.000000 -1.0
25% 3.001352e+06 8.000000 16.518374 74.645885 1.0 200.000000 1.000000 2.900000 6.000000 1.000000 0.0 -1.000000 -1.0
50% 1.869268e+07 26.000000 22.468629 77.106348 1.0 350.000000 1.000000 3.500000 35.000000 10.000000 0.0 -1.000000 -1.0
75% 1.887262e+07 11294.000000 26.752959 79.831641 1.0 600.000000 2.000000 3.900000 175.000000 69.000000 0.0 1.000000 -1.0
max 1.915979e+07 11354.000000 10000.000000 91.832769 1.0 30000.000000 4.000000 4.900000 42539.000000 17702.000000 0.0 1.000000 -1.0
In [56]:
full_summary_stats_categorical = zomato.describe(include=[object, 'category'])
full_summary_stats_categorical
Out[56]:
name establishment url address city locality locality_verbose cuisines timings currency highlights rating_text
count 55568 55568 55568 55568 55568 55568 55568 55568 54565 55568 55568 55568
unique 41100 27 55568 50658 99 3731 3910 9383 7740 1 31228 6
top Domino's Pizza Quick Bites https://www.zomato.com/agra/bikanervala-khanda... Laxman Jhula, Tapovan, Rishikesh Bangalore Civil Lines Gomti Nagar, Lucknow North Indian 11 AM to 11 PM Rs. ['Dinner', 'Takeaway Available', 'Lunch', 'Cas... Average
freq 399 14032 1 37 2247 752 274 4295 7012 55568 860 16323

Distribution Analysis¶

Analyzing Ratings Distribution¶

In [57]:
zomato['aggregate_rating'].hist(bins=20, edgecolor='black')
plt.title('Ratings Distribution')
plt.xlabel('Ratings')
plt.ylabel('Number of Restaurants')
plt.show()
In [58]:
sns.kdeplot(zomato['aggregate_rating'], fill=True)
plt.title("Ratings distribution")
plt.show()
  • The histogram shows two peaks around the ratings of 1 and between 3 and 4. This suggests that there are two prevalent groups of restaurants based on ratings — one group with low ratings and another group with higher ratings.
  • The majority of restaurants have ratings above 2.5, with a significant concentration between 3.0 and 4.0. This could indicate that most restaurants perform at a satisfactory level according to customer reviews
  • The first bar represents the count of restaurants with a rating of 0. This is unusually high compared to ratings of 1 or 2. There could be several explanations for this, such as a default or placeholder value for unrated restaurants, or it may include new restaurants that have not yet been rated.

Distrbution of Average Cost for two.¶

In [59]:
plt.figure(figsize=(10, 10))
plt.hist(zomato['average_cost_for_two'], bins=50, edgecolor='black', alpha=0.7)
plt.title('Distribution of Average Cost for Two')
plt.xlabel('Average Cost for Two')
plt.ylabel('Frequency')
plt.xlim(0, 5000)
plt.show()
In [60]:
sns.kdeplot(zomato['average_cost_for_two'], fill=True)
plt.xlim([0, 6000])
plt.xticks(range(0,6000,500))
plt.title("distribution of Average cost for two ")
plt.show()

Analyzing Price Range Distribution¶

In [61]:
price_counts = zomato['price_range'].value_counts()
sns.barplot(x=price_counts.index, y=price_counts.values)

for index, value in enumerate(price_counts.values):
    plt.text(index, value, str(value), color='black', ha="center")
    
plt.title('Distribution of Price Range')
plt.xlabel('Price Range')
plt.ylabel('Number of Restaurants')
plt.show()
  • Most Common Price Range: The most common price range for restaurants is "1", which has the highest number of restaurants. This indicates that a significant portion of the dataset's restaurants are positioned in the most affordable category.
  • Decreasing Frequency with Increasing Price Range: There's a clear descending order from the most common to the least common price range. As the price range increases from "1" to "4", the number of restaurants decreases.
  • Least Common Price Range: The least common price range is "4", which suggests that there are relatively few high-end, expensive restaurants compared to more affordable options.
  • Market Composition: The distribution suggests a market where budget-friendly dining options are abundant, mid-range restaurants are moderately available, and premium dining experiences are relatively rare.
  • Potential Business Insights: For entrepreneurs or restaurateurs, the chart might suggest that there is a saturated market for low-priced restaurants, potential room for growth in the mid-range, and a niche market for high-end establishments.
  • Consumer Behavior: The chart might reflect consumer behavior and spending patterns, indicating that a majority of diners may prefer or are looking for more economical dining options.
  • Data Representation: The y-axis scale and bar heights represent the absolute numbers of restaurants in each price range, but without the actual numbers, it is difficult to quantify the exact differences between the price ranges.
  • Economic Implications: The distribution might also provide insights into the economic status of the area from which the data is taken. An abundance of category "1" restaurants might indicate a more price-sensitive clientele or a region with lower average disposable income.
In [62]:
zomato['price_range'].value_counts().plot(kind='pie', autopct='%1.1f%%')
plt.title('Pie Chart of Price Range Distribution')
plt.ylabel('')  # Hide y-label as it's redundant in pie charts
plt.show()

Correlation Analysis¶

In [63]:
correlation_matrix = zomato[['aggregate_rating', 'average_cost_for_two', 'price_range', 'votes', 'photo_count']].corr()
In [64]:
plt.figure(figsize=(10, 8))
sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt=".2f", linewidths=.5)
plt.title('Correlation Matrix')
plt.show()

Cuisines¶

In [65]:
zomato['cuisines'].unique()
Out[65]:
array(['North Indian, South Indian, Mithai, Street Food, Desserts',
       'North Indian, Mughlai, Rolls, Chinese, Fast Food, Street Food',
       'Fast Food, Mithai', ...,
       'Street Food, Biryani, Chinese, Fast Food, North Indian, Mughlai',
       'North Indian, Chinese, Mexican, Italian, Thai, Continental',
       'North Indian, Lucknowi, Chinese'], dtype=object)
In [66]:
cuisines_expanded = zomato['cuisines'].str.split(', ').explode()
cuisine_counts = cuisines_expanded.value_counts()
In [67]:
top_cuisines = cuisine_counts.head(10)

plt.figure(figsize=(12, 8))
sns.barplot(x=top_cuisines.values, y=top_cuisines.index)
plt.title('Top 10 Cuisines by Number of Restaurants')
plt.xlabel('Number of Restaurants')
plt.ylabel('Cuisines')
plt.show()
In [68]:
establishment_counts = zomato['establishment'].value_counts()

# Plotting
plt.figure(figsize=(10, 6))
sns.barplot(x=establishment_counts.values, y=establishment_counts.index, palette='pastel')
plt.title('Distribution of Establishment')
plt.xlabel('Number of Establishments')
plt.ylabel('Types of Establishments')
plt.show()
  • 'aggregate_rating' has low positive correlations with 'average_cost_for_two'and 'price_range'. This could mean that higher-rated restaurants have a tendency to be slightly more expensive,, but the relationship is not very strong.
In [69]:
zomato['name'].value_counts()
Out[69]:
name
Domino's Pizza                           399
Cafe Coffee Day                          315
KFC                                      204
Baskin Robbins                           202
Keventers                                189
                                        ... 
Boulevard 69                               1
7-Eleven Restaurant                        1
Food station                               1
Silver Saloon - Taj Usha Kiran Palace      1
Geeta lodge                                1
Name: count, Length: 41100, dtype: int64

Restaurant¶

Some of the places have many chains. let seprate chains from single resturunt.

In [70]:
places = zomato['name'].value_counts()
In [71]:
places
Out[71]:
name
Domino's Pizza                           399
Cafe Coffee Day                          315
KFC                                      204
Baskin Robbins                           202
Keventers                                189
                                        ... 
Boulevard 69                               1
7-Eleven Restaurant                        1
Food station                               1
Silver Saloon - Taj Usha Kiran Palace      1
Geeta lodge                                1
Name: count, Length: 41100, dtype: int64
In [72]:
chains = places[places >= 2]
single = places[places == 1]
In [73]:
print("Total Number of Restaurants: ", zomato.shape[0])
print("Number of single places: ", single.count())
print("Number of chains : ", zomato.shape[0]-single.shape[0])
print("Percentage of Restaurants(chains) : ", np.round((zomato.shape[0] - single.shape[0]) / zomato.shape[0], 2)* 100, '%')

Total Number of Restaurants:  55568
Number of single places:  36210
Number of chains :  19358
Percentage of Restaurants(chains) :  35.0 %

Top Restaurant chains by No. of outlists¶

In [74]:
top10_chains = zomato["name"].value_counts()[:10].sort_values(ascending=True)
plt.figure(figsize=[11,7])
bars = plt.barh(top10_chains.index, top10_chains.values)
for bar in bars:
    plt.text(bar.get_width() + 3, bar.get_y() + bar.get_height()/2, f'{bar.get_width()}',
             va='center', color='black')

plt.xlabel("No. of outlets")
plt.title("Top 10 Restaurant Chains by number of outlits")

# Show the plot
plt.show()

Top Restaurant by Average rating¶

In [75]:
top10_chains = zomato["name"].value_counts()[zomato["name"].value_counts() > 4].index[:10]
top10_ratings = zomato[zomato["name"].isin(top10_chains)].groupby("name")["aggregate_rating"].mean().sort_values()

plt.figure(figsize=[11,7])
bars = plt.barh(top10_ratings.index, top10_ratings.values)

# Adding ratings to the bars
for i, v in enumerate(top10_ratings.values):
    plt.text(v + 0.01, i, f'{v:.2f}', color='black')

# Labeling and title
plt.xlabel("Average Rating")
plt.ylabel("Restaurant Chain")
plt.title("Top 10 Restaurant Chains by Average Rating)")

# Show the plot
plt.show()

Top 10 Cities by No. of restaurants¶

In [76]:
city_counts = zomato.groupby("city").count()["res_id"].nlargest(10).sort_values()
plt.figure(figsize=[11,7])
bars = plt.barh(city_counts.index, city_counts.values)
for bar in bars:
    plt.text(bar.get_width() + 20, bar.get_y() + bar.get_height()/2, f'{bar.get_width()}',
             va='center', color='black')
plt.xlabel("No. of Restaurants")
plt.title("Top 10 Cities by No. of Restaurants")
plt.show()

Number of Restaurants by Establishment¶

In [77]:
est_count = zomato.groupby("establishment").count()["res_id"].nlargest(10)
sns.barplot(y=est_count.index, x=est_count.values)
plt.xlabel("No. of Restaurants")
plt.title("Number of Restaurants (by Establishment)")
plt.show()

Competitive Analysis¶

In [85]:
zomato['cuisines']
Out[85]:
0         North Indian, South Indian, Mithai, Street Foo...
1         North Indian, Mughlai, Rolls, Chinese, Fast Fo...
2                                         Fast Food, Mithai
3                 Desserts, Bakery, Fast Food, South Indian
4                        North Indian, Continental, Italian
                                ...                        
211882                                            Ice Cream
211925                      Gujarati, North Indian, Chinese
211926                                Gujarati, Street Food
211940                                            Fast Food
211942                           Fast Food, Sandwich, Salad
Name: cuisines, Length: 55568, dtype: object
In [79]:
df_exploded = zomato.explode('cuisines')
df_exploded['aggregate_rating'] = pd.to_numeric(df_exploded['aggregate_rating'], errors='coerce')
df_exploded['average_cost_for_two'] = pd.to_numeric(df_exploded['average_cost_for_two'], errors='coerce')

competitors = df_exploded.groupby(['city', 'cuisines']).agg({
    'aggregate_rating': 'mean',
    'average_cost_for_two': 'mean',
    'name': 'size'
}).rename(columns={'name': 'restaurant_count'}).reset_index()

top_competitors = competitors.sort_values(by=['city', 'aggregate_rating', 'restaurant_count'], ascending=[True, False, False])
top_competitors = top_competitors.groupby('city').head(5)
In [80]:
for city, sub_df in top_competitors.groupby('city'):
    plt.figure(figsize=(10, 6))
    sns.barplot(data=sub_df, x='aggregate_rating', y='cuisines', hue='average_cost_for_two')
    plt.title(f'Top Competitors in {city}')
    plt.xlabel('average Rating')
    plt.ylabel('cuisine')
    plt.legend(title='average Cost for Two')
    plt.show()
In [81]:
strengths = top_competitors[(top_competitors['aggregate_rating'] >= 4) & (top_competitors['restaurant_count'] > top_competitors['restaurant_count'].median())]

# Weaknesses (lower ratings, fewer counts, high costs)
weaknesses = competitors[(competitors['aggregate_rating'] < 3.5) & (competitors['restaurant_count'] <= competitors['restaurant_count'].median()) & (competitors['average_cost_for_two'] > competitors['average_cost_for_two'].median())]

# Display strengths and weaknesses
print("Identified Strengths:")
print(strengths)
print("\nIdentified Weaknesses:")
print(weaknesses)

Identified Strengths:
                city                                           cuisines  \
1077        Amravati                         North Indian, South Indian   
2804          Bhopal  Street Food, South Indian, Fast Food, Desserts...   
2491          Bhopal  Bakery, Chinese, Continental, Pizza, Italian, ...   
3940         Chennai   North Indian, Mediterranean, Asian, Arabian, BBQ   
3463         Chennai             BBQ, North Indian, European, Beverages   
3855         Chennai                 Mexican, American, Tex-Mex, Burger   
5850   Greater Noida                       Burger, Fast Food, Beverages   
6423        Guwahati                                      Cafe, Italian   
6759         Gwalior                     North Indian, Biryani, Chinese   
7214       Hyderabad              European, Mediterranean, North Indian   
7020       Hyderabad                                           American   
8582           Jammu  Desserts, Fast Food, Chinese, North Indian, Mi...   
9470        Junagadh                                           Gujarati   
9475        Junagadh                                Ice Cream, Desserts   
9594          Kanpur                                 Fast Food, Italian   
10266       Kolhapur                                Ice Cream, Desserts   
11287        Lucknow                                      Modern Indian   
11757        Madurai                                  Burger, Fast Food   
11778        Madurai                Chinese, South Indian, North Indian   
11748        Madurai                          Biryani, Arabian, Chinese   
12106      Mangalore          Ice Cream, Desserts, Beverages, Fast Food   
13904         Nagpur                                Beverages, Desserts   
15092      New Delhi                               Italian, Finger Food   
15541          Noida              North Indian, Mughlai, Biryani, Rolls   
17779         Rajkot                         Sandwich, Burmese, Chinese   
17661         Rajkot                                  Burger, Fast Food   

       aggregate_rating  average_cost_for_two  restaurant_count  
1077           4.100000            500.000000                 2  
2804           4.650000            400.000000                 2  
2491           4.500000            500.000000                 3  
3940           4.900000           1400.000000                 3  
3463           4.820000           1400.000000                 5  
3855           4.800000           1400.000000                 2  
5850           4.050000            450.000000                 2  
6423           4.600000            800.000000                 2  
6759           4.500000            400.000000                 2  
7214           4.850000           1500.000000                 4  
7020           4.800000           1600.000000                 2  
8582           4.300000            300.000000                 2  
9470           4.150000            225.000000                 2  
9475           4.033333            200.000000                 3  
9594           4.550000            250.000000                 2  
10266          4.200000            275.000000                 2  
11287          4.750000           1300.000000                 2  
11757          4.450000            450.000000                 2  
11778          4.200000            200.000000                 2  
11748          4.150000            350.000000                 2  
12106          4.666667            283.333333                 3  
13904          4.650000            375.000000                 2  
15092          4.900000           2000.000000                 2  
15541          4.633333            516.666667                 3  
17779          4.300000            200.000000                 2  
17661          4.200000            400.000000                 2  

Identified Weaknesses:
           city                                           cuisines  \
1          Agra                       Asian, North Indian, Chinese   
2          Agra  Asian, North Indian, Chinese, South Indian, It...   
3          Agra  Asian, South Indian, Continental, Chinese, Nor...   
6          Agra         Bakery, Beverages, Cafe, Coffee, Fast Food   
14         Agra                                           Bar Food   
...         ...                                                ...   
21295  Zirakpur        North Indian, Chinese, Continental, Biryani   
21296  Zirakpur        North Indian, Chinese, Continental, Mexican   
21305  Zirakpur                 North Indian, Continental, Chinese   
21306  Zirakpur             North Indian, Continental, Finger Food   
21309  Zirakpur                              North Indian, Mughlai   

       aggregate_rating  average_cost_for_two  restaurant_count  
1                   0.0                1200.0                 1  
2                   0.0                1200.0                 1  
3                   0.0                2000.0                 1  
6                   3.3                 500.0                 1  
14                  0.0                1500.0                 1  
...                 ...                   ...               ...  
21295               3.2                1500.0                 1  
21296               3.4                1600.0                 1  
21305               2.7                 600.0                 1  
21306               3.2                1200.0                 1  
21309               2.4                 650.0                 1  

[1956 rows x 5 columns]